106 Spatial multi-omic characterization of tumor microenvironment heterogeneity in hepatocellular carcinoma
Bibliographic record
Abstract
Background Hepatocellular carcinoma (HCC) is a highly heterogeneous malignancy, demanding comprehensive multi-omic understanding of its spatial architecture to improve therapeutic strategies. Spatial transcriptomics with the Xenium™ platform enables mapping of hundreds to thousands of RNA targets in tissue sections. However, identification and validation of drug targets requires proteomic assessment to directly reveal the functional mechanisms underlying disease biology. Imaging Mass Cytometry™ (IMC™) is a spatial proteomic technology that utilizes cytometry by time-of-flight (on which CyTOF™ systems are based) to generate high-dimensional, spatially resolved protein expression data at subcellular resolution in tissue sections. Unlike traditional immunohistochemistry or immunofluorescence, IMC platforms use metal-tagged antibodies and laser ablation to simultaneously detect over 40 protein markers without spectral overlap or autofluorescence interference. Here we demonstrate the performance and insights gained from using IMC technology on tissue samples previously processed with the Xenium platform.Methods We profiled HCC FFPE tissue sections using a custom transcriptomic panel with Xenium v1 assay and then performed IMC using a 43-marker immuno-oncology-focused antibody panel on the same tissue section ( figure 1). We used IMC technology in parallel on non-Xenium processed control serial sections and compared performance. To combine the transcriptomic and proteomic datasets, we utilized Xenium Explorer software to co-register the datasets and observe overlay of transcriptomic and proteomic biomarkers.Results IMC technology alone and post-Xenium IMC generated data of similar quality, demonstrating highly consistent tumor and immune cell phenotyping capabilities in HCC ( figure 2). IMC technology alone and post-Xenium IMC similarly detect localization of macrophages (M1 and M2), neutrophils (CD66b), B cells (CD20), cytotoxic T cells (CD8) and T helper cells (CD4) in specific locations around the tissue. Xenium Explorer software permitted import of IMC data through a user-friendly co-registration algorithm that aligned stained nuclei from both datasets. Combining the datasets revealed the presence of subcategories of immune cells (T cells, B cells, macrophages) and their activation states. Additionally, detection of transcript and protein of multiple markers showcased discrepancies in spatial localization, highlighting the importance of validating transcriptomic data with proteomic assessment.Conclusions Here, we highlight the synergistic use of the Xenium platform for transcript detection and IMC technology for protein profiling on the same tissue section, facilitating an integrated understanding of tissue biology. This study provides a multi-omic view of HCC heterogeneity, offering insights into mechanisms of disease and development of potential therapeutic strategies.For Research Use Only. Not for use in diagnostic procedures.Abstract 106 Figure 1Combined workflow for spatial imaging of both RNA and protein on the same slide. The IMC staining protocol can be added onto the end of Xenium acquisition with an additional wash step prior to IMC staining. Slides can be imaged immediately or stored for later acquisitionAbstract 106 Figure 2Low and high abundance proteins were detected by IMC regardless of processing approach. IMC alone and post-Xenium IMC generate the same quality image, demonstrating highly consistent and reproducible data. For post-Xenium slides, the Xenium assay used was a custom panel using the Xenium v1 workflow
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".